Skip to content

CHANGELOG.md

Latest commit

 

History

History
815 lines (699 loc) · 37.3 KB

CHANGELOG.md

File metadata and controls

815 lines (699 loc) · 37.3 KB
Raw

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Added

  • Support for custom media types, new PointCloud media type, DatasetItem.media and .media_as(type) members (open-edge-platform#539)
  • [API] A way to request dataset and extractor media type with media_type (open-edge-platform#539)
  • BraTS format (import-only) (.npy and .nii.gz), new MultiframeImage media type (open-edge-platform#628)
  • Common Semantic Segmentation dataset format (import-only) (open-edge-platform#685)
  • An option to disable data/ prefix inclusion in YOLO export (open-edge-platform#689)
  • New command describe-downloads to print information about downloadable datasets (open-edge-platform#678)
  • Detection for Cityscapes format (open-edge-platform#680)
  • Maximum recursion --depth parameter for detect CLI command (open-edge-platform#680)
  • An option to save a single subset in the download command (open-edge-platform#697)
  • Common Super Resolution dataset format (import-only) (open-edge-platform#700)
  • Kinetics 400/600/700 dataset format (import-only) (open-edge-platform#706)
  • NYU Depth Dataset V2 format (import-only) (open-edge-platform#712)
  • Skeleton annotation type (cvat-ai#6, cvat-ai#14, cvat-ai#15)
  • Storing labels with the same name but with a different parent (cvat-ai#8)
  • Functions to work with plain polygons (COCO-style) - close_polygon, simplify_polygon (cvat-ai#39)
  • An option to specify scale factor in resize transform (cvat-ai#46)
  • Skeleton support in datumaro format (cvat-ai#47)
  • Support for Ultralytics YOLO formats (cvat-ai#50)
  • Support for Ultralytics YOLO Classification format (cvat-ai#59)
  • YOLO formats now merge default subset and train subset if both are present (cvat-ai#71)
  • Support for tracks in Ultralytics YOLO formats (cvat-ai#70)
  • [API] ImageFromBytes.save() now preserves image extension if no output extension is specified (cvat-ai#91)
  • [API] ImageFromBytes.save() now guarantees there will be no extra image encoding/decoding when possible (e.g. if input and output extension is the same) (cvat-ai#91)
  • [API] Added StreamingDatasetBase as a utility base type for user-created extractors with streaming support. Added support for using get_subset() from the source dataset in StreamDataset during iteration, when possible. (cvat-ai#97)

Changed

  • env.detect_dataset() now returns a list of detected formats at all recursion levels instead of just the lowest one (open-edge-platform#680)
  • Open Images: allowed to store annotations file in root path as well (open-edge-platform#680)
  • Improved parsing error messages in COCO, VOC and YOLO formats (open-edge-platform#684, open-edge-platform#686, open-edge-platform#687)
  • YOLO format now supports almost any subset names, except backup, names and classes (instead of just train and valid). The reserved names now raise an error on exporting. (open-edge-platform#688)
  • [CLI] Removed the --all flag in datum info, added the --json flag, added format and media type fields in the info command output (cvat-ai#5)
  • item id in MOT format (cvat-ai#17)
  • Annotation matching algorithm in datumaro.components.operations.match_segments() (cvat-ai#30)
  • Automatic detection of is_crowd parameter is disabled in segment_iou(), added a separate function argument (turned off by default) (cvat-ai#41)
  • Always use exif orientation info when loading images (cvat-ai#82)

Deprecated

  • --save-images is replaced with --save-media in CLI and converter API (open-edge-platform#539)
  • [API] image, point_cloud and related_images of DatasetItem are replaced with media and media_as(type) members and c-tor parameters (open-edge-platform#539)
  • [API] datumaro.util.annotation_util._get_bbox() is renamed into get_bbox() (cvat-ai#41)

Removed

  • KEEPS_SUBSETS_INTACT is not used anymore (cvat-ai#102)

Fixed

  • Collision between parents and names in LabelCategories
  • (cvat-ai#51)
  • Detection for LFW format (open-edge-platform#680)
  • Export of masks with background class with id != 0 in the VOC, KITTI and Cityscapes formats (cvat-ai#9, cvat-ai#16)
  • Missing comparison of the base class attributes in the Mask class (cvat-ai#28)
  • Image stats when no image info available for some images in the dataset (cvat-ai#29)
  • Incorrect writing of media field in the Datumaro format, when there are specific media fields (cvat-ai#34)
  • Added missing PointCloud media type in the datumaro module namespace (cvat-ai#34)
  • Incorrect computation of binary mask bbox (missed 1 pixel of the size) (cvat-ai#41)
  • Dataset.get() could ignore existing transforms in the dataset (cvat-ai#45)
  • Failing resize transform for RLE masks (cvat-ai#46)
  • Invalid handling of Mac OS special dirs in format detection (cvat-ai#88)
  • ICDAR exporter does not require index attributes anymore (cvat-ai#105)
  • MOTS exporter now works with RLE masks when NumPy 2.x is used (cvat-ai#132)
  • Cityscapes and KITTY exporters are now compatible with Pillow 13 (cvat-ai#136)

Security

  • TBD

21/02/2022 - Release v0.3

Added

Changed

  • Allowed direct file paths in datum import. Such sources are imported like when the rpath parameter is specified, however, only the selected path is copied into the project (open-edge-platform#555)
  • Improved stats performance, added new filtering parameters, image stats (unique, repeated) moved to the dataset section, removed mean and std from the dataset section (open-edge-platform#621)
  • Allowed Image creation from just size info (open-edge-platform#634)
  • Added image search in VOC XML-based subformats (open-edge-platform#634)
  • Added image path equality checks in simple merge, when applicable (open-edge-platform#634)
  • Supported saving box attributes when downloading the TFDS version of VOC (open-edge-platform#668)
  • Switched to a pyproject.toml-based build (open-edge-platform#671)

Deprecated

  • TBD

Removed

Fixed

Security

  • TBD

28/01/2022 - Release v0.2.3

Added

Changed

  • The pycocotools dependency lower bound is raised to 2.0.4. (open-edge-platform#449)
  • smooth_line from datumaro.util.annotation_util - the function is renamed to approximate_line and has updated interface (open-edge-platform#592)

Deprecated

  • Python 3.6 support

Removed

  • TBD

Fixed

Security

  • TBD

24/12/2021 - Release v0.2.2

Added

Changed

Deprecated

  • Using Image, ByteImage from datumaro.util.image - these classes are moved to datumaro.components.media (open-edge-platform#538)

Removed

  • Equality comparison support between datumaro.components.media.Image and numpy.ndarray (open-edge-platform#568)

Fixed

Security

  • TBD

16/11/2021 - Release v0.2.1

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

14/10/2021 - Release v0.2

Added

Changed

  • A project can contain and manage multiple datasets instead of a single one. CLI operations can be applied to the whole project, or to separate datasets. Datasets are modified inplace, by default (open-edge-platform#328)
  • CLI help for builtin plugins doesn't require project (open-edge-platform#328)
  • Annotation-related classes were moved into a new module, datumaro.components.annotation (open-edge-platform#439)
  • Rollback utilities replaced with Scope utilities (open-edge-platform#444)
  • The Project class from datumaro.components is changed completely (open-edge-platform#238)
  • diff and ediff are joined into a single diff CLI command (open-edge-platform#238)
  • Projects use new file layout, incompatible with old projects. An old project can be updated with datum project migrate (open-edge-platform#238)
  • Inheriting CliPlugin is not required in plugin classes (open-edge-platform#238)
  • Importers do not create Projects anymore and just return a list of extractor configurations (open-edge-platform#238)

Deprecated

  • TBD

Removed

Fixed

Security

  • TBD

24/08/2021 - Release v0.1.11

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

14/07/2021 - Release v0.1.10

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

03/06/2021 - Release v0.1.9

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

31/03/2021 - Release v0.1.8

Added

  • TBD

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

24/03/2021 - Release v0.1.7

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

03/02/2021 - Release v0.1.6.1 (hotfix)

Added

  • TBD

Changed

  • TBD

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

02/26/2021 - Release v0.1.6

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

01/23/2021 - Release v0.1.5

Added

Changed

Deprecated

Removed

  • TBD

Fixed

Security

  • TBD

12/10/2020 - Release v0.1.4

Added

Changed

Deprecated

  • TBD

Removed

  • TBD

Fixed

  • TBD

Security

  • TBD

10/28/2020 - Release v0.1.3

Added

Changed

  • TBD

Deprecated

  • TBD

Removed

  • TBD

Fixed

Security

  • TBD

10/05/2020 - Release v0.1.2

Added

  • ByteImage class to represent encoded images in memory and avoid recoding on save (open-edge-platform#27)

Changed

Deprecated

  • TBD

Removed

Fixed

Security

  • TBD

09/24/2020 - Release v0.1.1

Added

Changed

  • TBD

Deprecated

  • TBD

Removed

  • TBD

Fixed

  • TBD

Security

  • TBD

09/10/2020 - Release v0.1.0

Added

  • Initial release

Template

## [Unreleased]
### Added
- TBD

### Changed
- TBD

### Deprecated
- TBD

### Removed
- TBD

### Fixed
- TBD

### Security
- TBD